RESEARCH ARTICLE
Lower Mesophotic Coral Communities (60125 m Depth) of the Northern Great Barrier Reef and Coral Sea Norbert Englebert1,2,3*, Pim Bongaerts1,2, Paul R. Muir4, Kyra B. Hay1, Michel Pichon4, Ove Hoegh-Guldberg1,2
a1111111111 a1111111111 a1111111111 a1111111111 a1111111111
OPEN ACCESS Citation: Englebert N, Bongaerts P, Muir PR, Hay KB, Pichon M, Hoegh-Guldberg O (2017) Lower Mesophotic Coral Communities (60-125 m Depth) of the Northern Great Barrier Reef and Coral Sea. PLoS ONE 12(2): e0170336. doi:10.1371/journal. pone.0170336 Editor: Heather M. Patterson, Department of Agriculture and Water Resources, AUSTRALIA Received: July 11, 2016 Accepted: January 3, 2017 Published: February 1, 2017 Copyright: © 2017 Englebert et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: Specimens collected during the study are lodged and accessible at the Queensland Museum (www.qm.qld.gov.au) under accession numbers: G68156, G68392, G69875, G69880, G69908, G69943, G71132-G71158. Funding: This work was supported by the Catlin Group Limited and the Global Change Institute. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
1 Global Change Institute, The University of Queensland, St Lucia, QLD, Australia, 2 Australian Research Council Centre of Excellence for Coral Reef Studies, The University of Queensland, St Lucia, Queensland, Australia, 3 School of Biological Sciences, The University of Queensland, St Lucia, QLD, Australia, 4 Queensland Museum, Townsville, QLD, Australia * [email protected]
Abstract Mesophotic coral ecosystems in the Indo-Pacific remain relatively unexplored, particularly at lower mesophotic depths (60 m), despite their potentially large spatial extent. Here, we used a remotely operated vehicle to conduct a qualitative assessment of the zooxanthellate coral community at lower mesophotic depths (60–125 m) at 10 different locations in the Great Barrier Reef Marine Park and the Coral Sea Commonwealth Marine Reserve. Lower mesophotic coral communities were present at all 10 locations, with zooxanthellate scleractinian corals extending down to ~100 metres on walls and ~125 m on steep slopes. Lower mesophotic coral communities were most diverse in the 60–80 m zone, while at depths of 100 m the coral community consisted almost exclusively of the genus Leptoseris. Collections of coral specimens (n = 213) between 60 and 125 m depth confirmed the presence of at least 29 different species belonging to 18 genera, including several potential new species and geographic/depth range extensions. Overall, this study highlights that lower mesophotic coral ecosystems are likely to be ubiquitous features on the outer reefs of the Great Barrier Reef and atolls of the Coral Sea, and harbour a generic and species richness of corals that is much higher than thus far reported. Further research efforts are urgently required to better understand and manage these ecosystems as part of the Great Barrier Reef Marine Park and Coral Sea Commonwealth Marine Reserve.
Introduction Mesophotic coral ecosystems (MCEs) are found at intermediate depths (30 to ~100–150 m depth), and are characterized by the presence of scleractinian corals that (like their shallowwater counterparts) associate with dinoflagellates from the genus Symbiodinium (or “zooxanthellae”) [1]. Large areas of MCEs have been identified in the Indo-Pacific [2–4] and Western Atlantic (e.g. [5–7]), demonstrating that these ecosystems constitute a significant proportion of overall coral reef habitat [8]. However, most of the research efforts have concentrated on
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
1 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Competing Interests: The authors have declared that no competing interests exist.
upper mesophotic depths (30–60 m), given the potential of communities at these depths to provide a refuge against disturbances (e.g. warm-water bleaching and tropical storm damage) and act as a subsequent source of reproduction for damaged shallow reefs (given the substantial overlap in community structure) [9]. In contrast, studies on lower mesophotic depths (>60 m) remain relatively sparse (but see [10–13]), in part because their greater depth makes them harder to access. However, recent evidence from the Caribbean demonstrates that the lower mesophotic zone can host distinct, deep-specialist communities and represent a separate entity within the coral reef ecosystem [14]. For the world’s largest coral reef ecosystem, the Great Barrier Reef (GBR), studies have also focused primarily on the upper mesophotic zone, reporting diverse assemblages of both zooxanthellate scleractinian and octocorals [4, 15–17]. Yet, we still know surprisingly little about the presence and composition of lower MCEs, despite the original descriptions of lower mesophotic communities in this region dating back to the 1980s. Manned submersibles were used to confirm the presence of (zooxanthellate) coral-dominated communities at lower mesophotic depths at Myrmidon Reef and Ribbon Reef No. 5 (GBR; [18]) and Osprey Reef (Coral Sea; [19]). Since then, only a single survey has evaluated coral communities at lower mesophotic depths, reporting depth zonation patterns at four sites on the GBR [4, 20]. At these locations, either a dominance of azooxanthellate octocorals or a transitionary community containing both zooxanthellate and azooxanthellate taxa (60–75 m depth) was found to occur [20]. Nonetheless, there is extensive potential habitat (i.e. hard-substrate slopes in clear oceanic waters) for zooxanthellate scleractinian corals at lower mesophotic depths on the shelf edge reefs of the northern GBR (which extend over ~700 km) [21] and the steep walls of the numerous Coral Sea atolls (that rise out of deep oceanic waters) [15]. Yet, these lower mesophotic depths remain largely unexplored, in part due to the strict occupational safety constraints in scientific diving in this region. Consequently, confirmation of the extent of lower MCEs has been hampered by the requirement for manned or remotely controlled underwater vehicles. Zooxanthellate coral communities can extend to great depths, with zooxanthellate scleractinian corals observed in Hawaii and Johnston Atoll down to 153 and 165 m depth, respectively [11, 22]. Light availability is one of the key factors determining the vertical distribution of zooxanthellate scleractinian corals and is, to a large extent, predictable over depth due to the light attenuation characteristics of the water column [23]. In optically clear waters, light can penetrate to great depths with ~1% of the surface light irradiance reaching down to around 100 m in Hawaii [12]. The deepest observed limits for zooxanthellate corals are generally associated with locations with the highest optical water quality, which supports the hypothesis that light availability determines the depth distribution limits of zooxanthellate corals [23]. However, other factors such as temperature, sedimentation and reef structure might also play either a direct or indirect (by modulating the light availability) role in determining the lower depth limits of zooxanthellate corals [14, 24]; yet these factors and their interaction remain less well understood. While a recent study revealed that zooxanthellate corals can extend to at least 125 m depth on the GBR and Coral Sea atolls [25], it remains unclear to what extent this lower depth limit of zooxanthellate corals varies across north-east Australia, and how it relates to local environmental conditions. Based on the community structure data from elsewhere in the Indo-Pacific, the zooxanthellate scleractinian coral genus Leptoseris appears to be a key member of the lower mesophotic zone [23]. Additional coral genera Montipora, Porites and Pachyseris have been reported to occur in this depth zone in the Central Pacific Ocean [12, 26], yet were rarely found to be dominant or abundant below 60 m [but see 3]. Initial assessments of mesophotic coral biodiversity on the GBR [17, 27] revealed several new geographic and depth range extensions, however only a very small number of scleractinian corals from lower mesophotic depths (>60 m) were
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
2 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
recovered in these studies. Thus far, only 8 zooxanthellate scleractinian coral genera and 14 different species have been recorded at lower mesophotic depths ( 60 m) on the GBR and in the Coral Sea [15, 17–19, 27]. The low coral diversity reported to date could be due to the very limited sampling effort undertaken at lower mesophotic depths [27] and the lack of studies focusing specifically on the more diverse region of north-east Australia [28]. Despite the challenge of sampling at these depths, such studies are required to improve our understanding of the overall biodiversity on the GBR, and to determine the extent of unique biodiversity associated with the potentially extensive deep-water ecosystems. Here, we summarise qualitative assessments of lower mesophotic coral ecosystems carried out using a remotely operated vehicle (ROV) during five targeted research expeditions. In total, 15 different sites (representing 10 different reef locations) in the Great Barrier Reef Marine Park and the Coral Sea Commonwealth Marine Reserve were assessed, representing several degrees of latitude. The main aims of this study were to (1) identify the presence and lower depth limits of lower MCEs in north-east Australia; (2) assess broad patterns of zooxanthellate scleractinian coral community structure (at genus level) across the lower mesophotic depth gradient; and (3) evaluate whether these lower mesophotic depths harbour a low diversity of (zooxanthellate) scleractinian coral species.
Materials and Methods Study sites Lower mesophotic depths (60–125 m) were surveyed at a total of 15 different sites across 10 locations (Fig 1) in north-east Australian waters during five expeditions of the “Catlin Seaview Survey” carried out between September and December 2012, in November 2013, and in November-December 2014. Surveys and coral collections were conducted under permits from the Department of the Environment (018-CZRS-1207626-01, 018-RRRW-131031-01, AU-COM2012-151 and AU-COM2013-226) and from the Great Barrier Reef Marine Park Authority (G12/35281.1 and G14/37294.1). In the Great Barrier Reef Marine Park, we surveyed one site each at Raine Island, Great Detached Reef, Day Reef and Yonge Reef, and two sites each at Tijou Reef and Tydeman Reef (Fig 1). In the Coral Sea Commonwealth Marine Reserve, we surveyed four sites at Osprey Reef and one site each at Bougainville Reef, Holmes Reef and Flinders Reef (Table 1). Locations were chosen across a broad geographical range and, for many locations only a single site could be accessed due to sea-time limitations (full assessment of a site equated to a full day at sea). Individual sites were selected based on (and consequently biased towards) steep bathymetric profiles to allow for a relatively shallow anchorage with close-access to deep water for ROV operations.
Benthic video transects and analyses Benthic video transects were carried out using a vLBV-300 ROV (SeaBotix, San Diego USA) at 7 depth intervals (every 10 m) in the lower mesophotic zone between 60–120 m. Qualitative transects documenting the benthic community were conducted for 5–20 minutes at each depth, to assess coral community structure, bathymetric features and substratum composition. Shorter excursions were made at several sites to a maximum depth of ~130 m. The main ROV camera recorded in standard-definition (640 x 480 pixels) with an information overlay with depth, time and heading, and a second camera recorded in high-definition (1980 x 1024 pixels; used for analyses) and was time-synchronized with the main camera. A third camera with 10x optical zoom was frequently used to verify coral identifications in situ. For the bathymetric profile the following four categories were used: gentle slope (~10–30˚), moderate slope (~45˚), steep slope (~60˚) or wall (~80–90˚). The dominant substratum classes were scored sensu
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
3 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Fig 1. Map of survey locations. Survey locations in the Great Barrier Reef Marine Park (red dots) and the Coral Sea Commonwealth Marine Reserve (blue dots). Map produced with data files courtesy of Great Barrier Reef Marine Park Authority and “www.deepreef.org” under the Creative Commons Attribution Licence (CCAL) CC BY 4.0. doi:10.1371/journal.pone.0170336.g001
Bridge et al. [20]: sand/gravel/rubble (SGR), sediment-covered limestone (SCL) and limestone (L). Observed coral colonies were identified to genus level, as accurate species-level identification (which often requires microscopic examination of corallites for genera like Montipora) was usually not possible from the video footage. In addition, it was difficult at times to discriminate between some members of the genera Montipora and Porites, between members of the family Fungiidae, and between members of the genera previously ascribed to the family Pectiniidae (i.e. Echinophyllia, Mycedium, Oxypora and Echinomorpha). These were therefore classified as, respectively, Montipora/Porites, Fungiidae and “Pectiniidae”. Given that travelled distance was not consistent across depths and a proportion of the corals within the surveys could not be identified from video, representation of the dataset was simplified to presence/ absence data for each coral category (genus or family as described above). Colony sizes were categorized as small (50 cm), using the ROV manipulator as a size reference. Basic clustering and statistical analyses were carried out using the “vegan” package in R (v3.3.1). Hierarchical clustering (hclust) and analysis of variance
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
4 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Table 1. Details of surveyed locations including the site names, survey dates and coordinates (latitude and longitude). Reef location
Site name
Survey date
Latitude
Longitude
Raine Island
Raine
01-Dec-14
S11˚ 36’ 04.6"
E144˚ 02’ 55.4"
Great Detached Reef
Great Detached
11-Dec-12
S11˚ 42’ 14.8"
E144˚ 03’ 38.8"
Tijou Reef
Tijou 1
14-Dec-12
S13˚ 03’ 52.0"
E143˚ 57’ 02.4"
Tijou 2
29-Nov-14
S13˚ 03’ 07.1"
E143˚ 57’ 17.6" *
Tydeman 1
28-Nov-14
S13˚ 58’ 03.9"
E144˚ 31’ 52.6"
Tydeman 2
28-Nov-14
S13˚ 58’ 04.8"
E144˚ 30’ 11.8"
Day Reef
Day
18-Dec-12
S14˚ 28’ 27.5"
E145˚ 32’ 20.6"
Yonge Reef
Yonge
19-Dec-12
S14˚ 36’ 58.6"
E145˚ 38’ 12.6"
Osprey Reef
Osprey North Horn (NH)
29-Oct-12
S13˚ 48’ 27.0"
E146˚ 33’ 03.2"
Osprey Dutch Towers (DT)
26/27-Oct-12
S13˚ 49’ 13.2"
E146˚ 33’ 40.4"
Tydeman Reef
Osprey Bigeye Ledge (BL)
23-Oct-12
S13˚ 51’ 30.4"
E146˚ 33’ 43.0"
Osprey Halfway Wall (HW)
29-Oct-12
S13˚ 53’ 21.3"
E146˚ 33’ 18.1"
Bougainville Reef
Bougainville
23-Nov-13
S15˚ 30’ 09.9"
E147˚ 06’ 09.2"
Holmes Reef
Holmes
17/18-Sep-12
S16˚ 31’ 33.0"
E147˚ 48’ 23.9"
Flinders Reef
Flinders
21-Sep-12
S17˚ 41’ 41.0"
E148˚ 20’ 36.8"
* Approximate location. doi:10.1371/journal.pone.0170336.t001
(adonis) between transects was assessed using a Jaccard dissimilarity matrix of presence/ absence data (only transects with 10 observations were considered). The difference in generic diversity between the main depth range categories (60–80 m and 90–120 m) was assessed using a Mann-Whitney test.
Temperature measurements Temperature loggers (HOBO Pro v2 model U22-001; Onset Computer Corp., Bourne USA) were calibrated (at the Australian Institute of Marine Science, Townsville, Australia) and deployed at three sites on the northern GBR (Great Detached, Tijou 1 and Yonge) and at four study sites in the Coral Sea (Osprey Bigeye Ledge, Osprey Halfway Wall, Holmes and Flinders). Temperature loggers were deployed at depths of 10, 20 and 40 m (using SCUBA), and 60, 80 and 100 m (using the ROV) at each study site, and recorded water temperature at 5-minute intervals over a period of 22–48 hours from which averages were calculated.
Coral specimen collections Coral specimens were collected at several study sites as a means of checking video-based identifications and for identification to species level. A total of 213 coral colonies were sampled from 60 to 125 m depth using the ROV manipulator. Coral samples were tagged, photographed, sampled for DNA analyses and the remaining skeleton prepared for laboratory identification by bleaching for 24–72 hours in a sodium hypochlorite solution (house-hold bleach diluted in fresh-water) followed by rinsing in freshwater and drying in sunlight. Identifications of the bleached coral skeletal samples were conducted by Carden Wallace, Paul Muir and Michel Pichon at the Queensland Museum (QM; in Townsville) and representative specimens have been registered in the QM State Collection under numbers: G68156, G68392, G69875, G69880, G69908, G69943, G71132-G71158. These specimens and the matching video footage of the in situ coral colonies were used as a basis for genus identification based solely on video footage.
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
5 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Results Bathymetric profiles, substrate composition and temperatures Great Barrier Reef–Two distinct bathymetric profiles were observed (Fig 2) on the surveyed outer reef locations on the northern GBR: (1) a plateau consisting of sand, gravel and rubble dominated by Halimeda macroalgae (gently sloping; ~10–30˚) turning into a steep slope around 70–80 m (Yonge and Day) or near-vertical wall around 90 m depth (Great Detached), and (2) a moderate slope (~45˚) of limestone covered in varying degrees with sediment that became steeper with depth and eventually turned into a near-vertical wall between 80–100 m depth (Raine, Tijou, Tydeman). Halimeda macroalgae were only found to be dominant in the
Yonge
Day
Tydeman 1
Tydeman 2
Tijou 1
Tijou 2
Bougainville
Osprey NH
Osprey DT
Osprey BL
Gr.Detached
Raine
60 m
70 m
80 m
90 m
100 m
110 m
120 m
125 m
Flinders
125 m
Holmes
Osprey HW Leptoseris Pectiniidae Pachyseris Montip./Poritis Goniopora Fungiidae Parascolymia Cynarina Acropora
60 m
70 m
80 m
Gentle slope Moderate Slope
90 m Steep slope Wall
100 m
SGR
110 m
SCL SCL dominated
120 m
125 m
115 m
L dominated
Fig 2. Presence/absence diagram of coral genera over depth across 15 survey sites on the Great Barrier Reef (top row) and Coral Sea (bottom row). Colours indicate different growth forms (blues = plating, greens = encrusting/plating, browns = solitary/free-living and yellow = branching) and shades indicate different taxonomic groups. For the benthic substrate, steepness is indicated with different shapes and dominant substrate type is indicated by colour (SGR = sand/gravel/rubble; SCL = sediment-covered limestone; SCL dominated = sedimentcovered limestone dominated with some exposed limestone present; L dominated = exposed limestone dominated, but some sediment-covered limestone present). Grey horizontal “snow” means no transect performed at those depths. doi:10.1371/journal.pone.0170336.g002
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
6 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Fig 3. Video stills from benthic video transects. (a) Green macroalgae (genus Cladophora) covering a Montipora colony at Tydeman Reef at 70 m depth. (b) Monostand of fragmented Goniopora encountered at Osprey Reef HW around 70 m depth, interrupted by small sand patches and sediment-covered limestone. Close-up of colony inserted in left bottom corner. (c) Deep Leptoseris community consisting of small (10 different genera, which is
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
10 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Table 2. Zooxanthellate scleractinian coral species collected at lower mesophotic depths (60–125 m) on the Great Barrier Reef (GBR) and in the Coral Sea (CS). Species
Max. collection depth
Collection location(s)
Acropora aculeus1
60 m
Osprey (CS)
Acropora echinata1
60 m
Osprey (CS)
Acropora cf. granulosa1
73 m
Day (GBR)
Acropora speciosa1
60 m
Osprey (CS)
Alveopora cf. tizardi
60 m
Osprey (CS) Osprey (CS)
2
Bernardpora stutchburyi
60 m
Coscinaraea sp.
60 m
Osprey (CS)
Craterastrea levis3
60 m
Osprey (CS)
Cycloseris vaughani2
60 m
Osprey (CS)
Cynarina lacrymalis4
60 m
Tydeman (GBR)
Cyphastrea sp.
60 m
Osprey (CS)
cf. Echinophyllia sp.
101 m
Yonge (GBR)
Echinomorpha nishihirai
82 m
Osprey (CS) Tydeman (GBR)
Leptoseris fragilis
80 m
Leptoseris cf fragilis
124 m
Day (GBR)
Leptoseris glabra2
60 m
Tijou (GBR), Osprey (CS)
Leptoseris hawaiiensis
125 m
Bougainville (CS)
Leptoseris mycetoseroides2
60 m
Osprey (CS)
Leptoseris scabra
80 m
Holmes (CS), Osprey (CS)
Lithophyllon undulatum2
60 m
Osprey (CS)
Montipora aequituberculata2
60 m
Osprey (CS)
Montipora millepora
60 m
Osprey (CS)
Mycedium elephantotus
60 m
Tijou (GBR), Osprey (CS)
Oxypora glabra2
60 m
Tijou (GBR), Flinders (CS)
Oxypora lacera2
64 m
Osprey (CS)
Pachyseris speciosa
80 m
Holmes (CS), Osprey (CS)
Parascolymia vitiensis2, 4
60 m
Tydeman (GBR)
Podabacia crustacea2
60 m
Osprey (CS)
Porites rus2
65 m
Osprey (CS)
Seriatopora hystrix
60 m
Osprey (CS)
Stylocoeniella guenther2
60 m
Osprey (CS)
1
Previously recorded in Muir et al. [17].
2
New depth record for Great Barrier Reef / Coral Sea.
3
New record of occurrence for Australia. Species described from video observations.
4
doi:10.1371/journal.pone.0170336.t002
broadly similar to other parts of the Indo-Pacific (e.g. [3, 12]). Diversity was highest between 60 and 80 m depth, with the lowest reaches of the mesophotic zone (100–120 m) dominated almost exclusively by the genus Leptoseris. These observations are again similar to other parts of the Indo-Pacific [23] and reflect the unique ability of this genus to withstand extremely low light conditions [31–34]. Several members of the family Fungiidae were observed between 100 and 120 m depth in the Coral Sea, and while alive and well-pigmented, it cannot be excluded that these have recently tumbled down the slope from shallower depths. Besides the zooxanthellate scleractinian coral genera that are dominant in the lower mesophotic zone, colonies belonging to the genera Acropora and Goniopora were occasionally observed. Although Acropora is generally considered to be a “shallow-water” genus, a recent study by Muir et al. [17]
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
11 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
demonstrated that Acropora spp. can be relatively abundant in the upper mesophotic zone (
Lower Mesophotic Coral Communities (60125 m Depth) of the Northern Great Barrier Reef and Coral Sea Norbert Englebert1,2,3*, Pim Bongaerts1,2, Paul R. Muir4, Kyra B. Hay1, Michel Pichon4, Ove Hoegh-Guldberg1,2
a1111111111 a1111111111 a1111111111 a1111111111 a1111111111
OPEN ACCESS Citation: Englebert N, Bongaerts P, Muir PR, Hay KB, Pichon M, Hoegh-Guldberg O (2017) Lower Mesophotic Coral Communities (60-125 m Depth) of the Northern Great Barrier Reef and Coral Sea. PLoS ONE 12(2): e0170336. doi:10.1371/journal. pone.0170336 Editor: Heather M. Patterson, Department of Agriculture and Water Resources, AUSTRALIA Received: July 11, 2016 Accepted: January 3, 2017 Published: February 1, 2017 Copyright: © 2017 Englebert et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: Specimens collected during the study are lodged and accessible at the Queensland Museum (www.qm.qld.gov.au) under accession numbers: G68156, G68392, G69875, G69880, G69908, G69943, G71132-G71158. Funding: This work was supported by the Catlin Group Limited and the Global Change Institute. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
1 Global Change Institute, The University of Queensland, St Lucia, QLD, Australia, 2 Australian Research Council Centre of Excellence for Coral Reef Studies, The University of Queensland, St Lucia, Queensland, Australia, 3 School of Biological Sciences, The University of Queensland, St Lucia, QLD, Australia, 4 Queensland Museum, Townsville, QLD, Australia * [email protected]
Abstract Mesophotic coral ecosystems in the Indo-Pacific remain relatively unexplored, particularly at lower mesophotic depths (60 m), despite their potentially large spatial extent. Here, we used a remotely operated vehicle to conduct a qualitative assessment of the zooxanthellate coral community at lower mesophotic depths (60–125 m) at 10 different locations in the Great Barrier Reef Marine Park and the Coral Sea Commonwealth Marine Reserve. Lower mesophotic coral communities were present at all 10 locations, with zooxanthellate scleractinian corals extending down to ~100 metres on walls and ~125 m on steep slopes. Lower mesophotic coral communities were most diverse in the 60–80 m zone, while at depths of 100 m the coral community consisted almost exclusively of the genus Leptoseris. Collections of coral specimens (n = 213) between 60 and 125 m depth confirmed the presence of at least 29 different species belonging to 18 genera, including several potential new species and geographic/depth range extensions. Overall, this study highlights that lower mesophotic coral ecosystems are likely to be ubiquitous features on the outer reefs of the Great Barrier Reef and atolls of the Coral Sea, and harbour a generic and species richness of corals that is much higher than thus far reported. Further research efforts are urgently required to better understand and manage these ecosystems as part of the Great Barrier Reef Marine Park and Coral Sea Commonwealth Marine Reserve.
Introduction Mesophotic coral ecosystems (MCEs) are found at intermediate depths (30 to ~100–150 m depth), and are characterized by the presence of scleractinian corals that (like their shallowwater counterparts) associate with dinoflagellates from the genus Symbiodinium (or “zooxanthellae”) [1]. Large areas of MCEs have been identified in the Indo-Pacific [2–4] and Western Atlantic (e.g. [5–7]), demonstrating that these ecosystems constitute a significant proportion of overall coral reef habitat [8]. However, most of the research efforts have concentrated on
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
1 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Competing Interests: The authors have declared that no competing interests exist.
upper mesophotic depths (30–60 m), given the potential of communities at these depths to provide a refuge against disturbances (e.g. warm-water bleaching and tropical storm damage) and act as a subsequent source of reproduction for damaged shallow reefs (given the substantial overlap in community structure) [9]. In contrast, studies on lower mesophotic depths (>60 m) remain relatively sparse (but see [10–13]), in part because their greater depth makes them harder to access. However, recent evidence from the Caribbean demonstrates that the lower mesophotic zone can host distinct, deep-specialist communities and represent a separate entity within the coral reef ecosystem [14]. For the world’s largest coral reef ecosystem, the Great Barrier Reef (GBR), studies have also focused primarily on the upper mesophotic zone, reporting diverse assemblages of both zooxanthellate scleractinian and octocorals [4, 15–17]. Yet, we still know surprisingly little about the presence and composition of lower MCEs, despite the original descriptions of lower mesophotic communities in this region dating back to the 1980s. Manned submersibles were used to confirm the presence of (zooxanthellate) coral-dominated communities at lower mesophotic depths at Myrmidon Reef and Ribbon Reef No. 5 (GBR; [18]) and Osprey Reef (Coral Sea; [19]). Since then, only a single survey has evaluated coral communities at lower mesophotic depths, reporting depth zonation patterns at four sites on the GBR [4, 20]. At these locations, either a dominance of azooxanthellate octocorals or a transitionary community containing both zooxanthellate and azooxanthellate taxa (60–75 m depth) was found to occur [20]. Nonetheless, there is extensive potential habitat (i.e. hard-substrate slopes in clear oceanic waters) for zooxanthellate scleractinian corals at lower mesophotic depths on the shelf edge reefs of the northern GBR (which extend over ~700 km) [21] and the steep walls of the numerous Coral Sea atolls (that rise out of deep oceanic waters) [15]. Yet, these lower mesophotic depths remain largely unexplored, in part due to the strict occupational safety constraints in scientific diving in this region. Consequently, confirmation of the extent of lower MCEs has been hampered by the requirement for manned or remotely controlled underwater vehicles. Zooxanthellate coral communities can extend to great depths, with zooxanthellate scleractinian corals observed in Hawaii and Johnston Atoll down to 153 and 165 m depth, respectively [11, 22]. Light availability is one of the key factors determining the vertical distribution of zooxanthellate scleractinian corals and is, to a large extent, predictable over depth due to the light attenuation characteristics of the water column [23]. In optically clear waters, light can penetrate to great depths with ~1% of the surface light irradiance reaching down to around 100 m in Hawaii [12]. The deepest observed limits for zooxanthellate corals are generally associated with locations with the highest optical water quality, which supports the hypothesis that light availability determines the depth distribution limits of zooxanthellate corals [23]. However, other factors such as temperature, sedimentation and reef structure might also play either a direct or indirect (by modulating the light availability) role in determining the lower depth limits of zooxanthellate corals [14, 24]; yet these factors and their interaction remain less well understood. While a recent study revealed that zooxanthellate corals can extend to at least 125 m depth on the GBR and Coral Sea atolls [25], it remains unclear to what extent this lower depth limit of zooxanthellate corals varies across north-east Australia, and how it relates to local environmental conditions. Based on the community structure data from elsewhere in the Indo-Pacific, the zooxanthellate scleractinian coral genus Leptoseris appears to be a key member of the lower mesophotic zone [23]. Additional coral genera Montipora, Porites and Pachyseris have been reported to occur in this depth zone in the Central Pacific Ocean [12, 26], yet were rarely found to be dominant or abundant below 60 m [but see 3]. Initial assessments of mesophotic coral biodiversity on the GBR [17, 27] revealed several new geographic and depth range extensions, however only a very small number of scleractinian corals from lower mesophotic depths (>60 m) were
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
2 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
recovered in these studies. Thus far, only 8 zooxanthellate scleractinian coral genera and 14 different species have been recorded at lower mesophotic depths ( 60 m) on the GBR and in the Coral Sea [15, 17–19, 27]. The low coral diversity reported to date could be due to the very limited sampling effort undertaken at lower mesophotic depths [27] and the lack of studies focusing specifically on the more diverse region of north-east Australia [28]. Despite the challenge of sampling at these depths, such studies are required to improve our understanding of the overall biodiversity on the GBR, and to determine the extent of unique biodiversity associated with the potentially extensive deep-water ecosystems. Here, we summarise qualitative assessments of lower mesophotic coral ecosystems carried out using a remotely operated vehicle (ROV) during five targeted research expeditions. In total, 15 different sites (representing 10 different reef locations) in the Great Barrier Reef Marine Park and the Coral Sea Commonwealth Marine Reserve were assessed, representing several degrees of latitude. The main aims of this study were to (1) identify the presence and lower depth limits of lower MCEs in north-east Australia; (2) assess broad patterns of zooxanthellate scleractinian coral community structure (at genus level) across the lower mesophotic depth gradient; and (3) evaluate whether these lower mesophotic depths harbour a low diversity of (zooxanthellate) scleractinian coral species.
Materials and Methods Study sites Lower mesophotic depths (60–125 m) were surveyed at a total of 15 different sites across 10 locations (Fig 1) in north-east Australian waters during five expeditions of the “Catlin Seaview Survey” carried out between September and December 2012, in November 2013, and in November-December 2014. Surveys and coral collections were conducted under permits from the Department of the Environment (018-CZRS-1207626-01, 018-RRRW-131031-01, AU-COM2012-151 and AU-COM2013-226) and from the Great Barrier Reef Marine Park Authority (G12/35281.1 and G14/37294.1). In the Great Barrier Reef Marine Park, we surveyed one site each at Raine Island, Great Detached Reef, Day Reef and Yonge Reef, and two sites each at Tijou Reef and Tydeman Reef (Fig 1). In the Coral Sea Commonwealth Marine Reserve, we surveyed four sites at Osprey Reef and one site each at Bougainville Reef, Holmes Reef and Flinders Reef (Table 1). Locations were chosen across a broad geographical range and, for many locations only a single site could be accessed due to sea-time limitations (full assessment of a site equated to a full day at sea). Individual sites were selected based on (and consequently biased towards) steep bathymetric profiles to allow for a relatively shallow anchorage with close-access to deep water for ROV operations.
Benthic video transects and analyses Benthic video transects were carried out using a vLBV-300 ROV (SeaBotix, San Diego USA) at 7 depth intervals (every 10 m) in the lower mesophotic zone between 60–120 m. Qualitative transects documenting the benthic community were conducted for 5–20 minutes at each depth, to assess coral community structure, bathymetric features and substratum composition. Shorter excursions were made at several sites to a maximum depth of ~130 m. The main ROV camera recorded in standard-definition (640 x 480 pixels) with an information overlay with depth, time and heading, and a second camera recorded in high-definition (1980 x 1024 pixels; used for analyses) and was time-synchronized with the main camera. A third camera with 10x optical zoom was frequently used to verify coral identifications in situ. For the bathymetric profile the following four categories were used: gentle slope (~10–30˚), moderate slope (~45˚), steep slope (~60˚) or wall (~80–90˚). The dominant substratum classes were scored sensu
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
3 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Fig 1. Map of survey locations. Survey locations in the Great Barrier Reef Marine Park (red dots) and the Coral Sea Commonwealth Marine Reserve (blue dots). Map produced with data files courtesy of Great Barrier Reef Marine Park Authority and “www.deepreef.org” under the Creative Commons Attribution Licence (CCAL) CC BY 4.0. doi:10.1371/journal.pone.0170336.g001
Bridge et al. [20]: sand/gravel/rubble (SGR), sediment-covered limestone (SCL) and limestone (L). Observed coral colonies were identified to genus level, as accurate species-level identification (which often requires microscopic examination of corallites for genera like Montipora) was usually not possible from the video footage. In addition, it was difficult at times to discriminate between some members of the genera Montipora and Porites, between members of the family Fungiidae, and between members of the genera previously ascribed to the family Pectiniidae (i.e. Echinophyllia, Mycedium, Oxypora and Echinomorpha). These were therefore classified as, respectively, Montipora/Porites, Fungiidae and “Pectiniidae”. Given that travelled distance was not consistent across depths and a proportion of the corals within the surveys could not be identified from video, representation of the dataset was simplified to presence/ absence data for each coral category (genus or family as described above). Colony sizes were categorized as small (50 cm), using the ROV manipulator as a size reference. Basic clustering and statistical analyses were carried out using the “vegan” package in R (v3.3.1). Hierarchical clustering (hclust) and analysis of variance
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
4 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Table 1. Details of surveyed locations including the site names, survey dates and coordinates (latitude and longitude). Reef location
Site name
Survey date
Latitude
Longitude
Raine Island
Raine
01-Dec-14
S11˚ 36’ 04.6"
E144˚ 02’ 55.4"
Great Detached Reef
Great Detached
11-Dec-12
S11˚ 42’ 14.8"
E144˚ 03’ 38.8"
Tijou Reef
Tijou 1
14-Dec-12
S13˚ 03’ 52.0"
E143˚ 57’ 02.4"
Tijou 2
29-Nov-14
S13˚ 03’ 07.1"
E143˚ 57’ 17.6" *
Tydeman 1
28-Nov-14
S13˚ 58’ 03.9"
E144˚ 31’ 52.6"
Tydeman 2
28-Nov-14
S13˚ 58’ 04.8"
E144˚ 30’ 11.8"
Day Reef
Day
18-Dec-12
S14˚ 28’ 27.5"
E145˚ 32’ 20.6"
Yonge Reef
Yonge
19-Dec-12
S14˚ 36’ 58.6"
E145˚ 38’ 12.6"
Osprey Reef
Osprey North Horn (NH)
29-Oct-12
S13˚ 48’ 27.0"
E146˚ 33’ 03.2"
Osprey Dutch Towers (DT)
26/27-Oct-12
S13˚ 49’ 13.2"
E146˚ 33’ 40.4"
Tydeman Reef
Osprey Bigeye Ledge (BL)
23-Oct-12
S13˚ 51’ 30.4"
E146˚ 33’ 43.0"
Osprey Halfway Wall (HW)
29-Oct-12
S13˚ 53’ 21.3"
E146˚ 33’ 18.1"
Bougainville Reef
Bougainville
23-Nov-13
S15˚ 30’ 09.9"
E147˚ 06’ 09.2"
Holmes Reef
Holmes
17/18-Sep-12
S16˚ 31’ 33.0"
E147˚ 48’ 23.9"
Flinders Reef
Flinders
21-Sep-12
S17˚ 41’ 41.0"
E148˚ 20’ 36.8"
* Approximate location. doi:10.1371/journal.pone.0170336.t001
(adonis) between transects was assessed using a Jaccard dissimilarity matrix of presence/ absence data (only transects with 10 observations were considered). The difference in generic diversity between the main depth range categories (60–80 m and 90–120 m) was assessed using a Mann-Whitney test.
Temperature measurements Temperature loggers (HOBO Pro v2 model U22-001; Onset Computer Corp., Bourne USA) were calibrated (at the Australian Institute of Marine Science, Townsville, Australia) and deployed at three sites on the northern GBR (Great Detached, Tijou 1 and Yonge) and at four study sites in the Coral Sea (Osprey Bigeye Ledge, Osprey Halfway Wall, Holmes and Flinders). Temperature loggers were deployed at depths of 10, 20 and 40 m (using SCUBA), and 60, 80 and 100 m (using the ROV) at each study site, and recorded water temperature at 5-minute intervals over a period of 22–48 hours from which averages were calculated.
Coral specimen collections Coral specimens were collected at several study sites as a means of checking video-based identifications and for identification to species level. A total of 213 coral colonies were sampled from 60 to 125 m depth using the ROV manipulator. Coral samples were tagged, photographed, sampled for DNA analyses and the remaining skeleton prepared for laboratory identification by bleaching for 24–72 hours in a sodium hypochlorite solution (house-hold bleach diluted in fresh-water) followed by rinsing in freshwater and drying in sunlight. Identifications of the bleached coral skeletal samples were conducted by Carden Wallace, Paul Muir and Michel Pichon at the Queensland Museum (QM; in Townsville) and representative specimens have been registered in the QM State Collection under numbers: G68156, G68392, G69875, G69880, G69908, G69943, G71132-G71158. These specimens and the matching video footage of the in situ coral colonies were used as a basis for genus identification based solely on video footage.
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
5 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Results Bathymetric profiles, substrate composition and temperatures Great Barrier Reef–Two distinct bathymetric profiles were observed (Fig 2) on the surveyed outer reef locations on the northern GBR: (1) a plateau consisting of sand, gravel and rubble dominated by Halimeda macroalgae (gently sloping; ~10–30˚) turning into a steep slope around 70–80 m (Yonge and Day) or near-vertical wall around 90 m depth (Great Detached), and (2) a moderate slope (~45˚) of limestone covered in varying degrees with sediment that became steeper with depth and eventually turned into a near-vertical wall between 80–100 m depth (Raine, Tijou, Tydeman). Halimeda macroalgae were only found to be dominant in the
Yonge
Day
Tydeman 1
Tydeman 2
Tijou 1
Tijou 2
Bougainville
Osprey NH
Osprey DT
Osprey BL
Gr.Detached
Raine
60 m
70 m
80 m
90 m
100 m
110 m
120 m
125 m
Flinders
125 m
Holmes
Osprey HW Leptoseris Pectiniidae Pachyseris Montip./Poritis Goniopora Fungiidae Parascolymia Cynarina Acropora
60 m
70 m
80 m
Gentle slope Moderate Slope
90 m Steep slope Wall
100 m
SGR
110 m
SCL SCL dominated
120 m
125 m
115 m
L dominated
Fig 2. Presence/absence diagram of coral genera over depth across 15 survey sites on the Great Barrier Reef (top row) and Coral Sea (bottom row). Colours indicate different growth forms (blues = plating, greens = encrusting/plating, browns = solitary/free-living and yellow = branching) and shades indicate different taxonomic groups. For the benthic substrate, steepness is indicated with different shapes and dominant substrate type is indicated by colour (SGR = sand/gravel/rubble; SCL = sediment-covered limestone; SCL dominated = sedimentcovered limestone dominated with some exposed limestone present; L dominated = exposed limestone dominated, but some sediment-covered limestone present). Grey horizontal “snow” means no transect performed at those depths. doi:10.1371/journal.pone.0170336.g002
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
6 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Fig 3. Video stills from benthic video transects. (a) Green macroalgae (genus Cladophora) covering a Montipora colony at Tydeman Reef at 70 m depth. (b) Monostand of fragmented Goniopora encountered at Osprey Reef HW around 70 m depth, interrupted by small sand patches and sediment-covered limestone. Close-up of colony inserted in left bottom corner. (c) Deep Leptoseris community consisting of small (10 different genera, which is
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
10 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
Table 2. Zooxanthellate scleractinian coral species collected at lower mesophotic depths (60–125 m) on the Great Barrier Reef (GBR) and in the Coral Sea (CS). Species
Max. collection depth
Collection location(s)
Acropora aculeus1
60 m
Osprey (CS)
Acropora echinata1
60 m
Osprey (CS)
Acropora cf. granulosa1
73 m
Day (GBR)
Acropora speciosa1
60 m
Osprey (CS)
Alveopora cf. tizardi
60 m
Osprey (CS) Osprey (CS)
2
Bernardpora stutchburyi
60 m
Coscinaraea sp.
60 m
Osprey (CS)
Craterastrea levis3
60 m
Osprey (CS)
Cycloseris vaughani2
60 m
Osprey (CS)
Cynarina lacrymalis4
60 m
Tydeman (GBR)
Cyphastrea sp.
60 m
Osprey (CS)
cf. Echinophyllia sp.
101 m
Yonge (GBR)
Echinomorpha nishihirai
82 m
Osprey (CS) Tydeman (GBR)
Leptoseris fragilis
80 m
Leptoseris cf fragilis
124 m
Day (GBR)
Leptoseris glabra2
60 m
Tijou (GBR), Osprey (CS)
Leptoseris hawaiiensis
125 m
Bougainville (CS)
Leptoseris mycetoseroides2
60 m
Osprey (CS)
Leptoseris scabra
80 m
Holmes (CS), Osprey (CS)
Lithophyllon undulatum2
60 m
Osprey (CS)
Montipora aequituberculata2
60 m
Osprey (CS)
Montipora millepora
60 m
Osprey (CS)
Mycedium elephantotus
60 m
Tijou (GBR), Osprey (CS)
Oxypora glabra2
60 m
Tijou (GBR), Flinders (CS)
Oxypora lacera2
64 m
Osprey (CS)
Pachyseris speciosa
80 m
Holmes (CS), Osprey (CS)
Parascolymia vitiensis2, 4
60 m
Tydeman (GBR)
Podabacia crustacea2
60 m
Osprey (CS)
Porites rus2
65 m
Osprey (CS)
Seriatopora hystrix
60 m
Osprey (CS)
Stylocoeniella guenther2
60 m
Osprey (CS)
1
Previously recorded in Muir et al. [17].
2
New depth record for Great Barrier Reef / Coral Sea.
3
New record of occurrence for Australia. Species described from video observations.
4
doi:10.1371/journal.pone.0170336.t002
broadly similar to other parts of the Indo-Pacific (e.g. [3, 12]). Diversity was highest between 60 and 80 m depth, with the lowest reaches of the mesophotic zone (100–120 m) dominated almost exclusively by the genus Leptoseris. These observations are again similar to other parts of the Indo-Pacific [23] and reflect the unique ability of this genus to withstand extremely low light conditions [31–34]. Several members of the family Fungiidae were observed between 100 and 120 m depth in the Coral Sea, and while alive and well-pigmented, it cannot be excluded that these have recently tumbled down the slope from shallower depths. Besides the zooxanthellate scleractinian coral genera that are dominant in the lower mesophotic zone, colonies belonging to the genera Acropora and Goniopora were occasionally observed. Although Acropora is generally considered to be a “shallow-water” genus, a recent study by Muir et al. [17]
PLOS ONE | DOI:10.1371/journal.pone.0170336 February 1, 2017
11 / 16
Lower Mesophotic Coral Communities of the Northern GBR and Coral Sea
demonstrated that Acropora spp. can be relatively abundant in the upper mesophotic zone (
